API reference

Before you start#

You need an API token.

1. In Reseek, open Settings, then API Tokens.
2. Create a token and copy it. Keep it somewhere safe.

A token can have two permissions:

• Read lets a token see your library, tags, stats, money, reminders and quizzes.
• Write lets a token save, change and delete items, and create or change money entries, reminders and quizzes.

Tokens can be set to stop working after 7, 30, 90 or 365 days.

Address and headers#

What	Value
API base URL	https://reseek.net/api
Auth header	Authorization: Bearer YOUR_API_TOKEN
Recommended header	Accept: application/json

Check your token#

Run this first. If it succeeds, your token and header are set up correctly.

curl "https://reseek.net/api/user" \
  -H "Authorization: Bearer YOUR_API_TOKEN" \
  -H "Accept: application/json"

Limits#

• Each token can make up to 120 requests a minute. Past that you get a 429 response with a Retry-After header.
• Low-level scrape, OCR, AI chat and dashboard utility endpoints are not available through API tokens.
• Money endpoints keep records only. They do not move real money.

The endpoints below match the tools that AI apps get through MCP. See Connect a coding agent.

Your account#

GET /api/user#

Returns your profile and what the current token can do (abilities and expiry).

No body required. Works with any valid token.

Library#

POST /api/library#

Search or browse saved library items. Sentence queries are translated and converted into search keywords automatically.

Body: optional query, type, date filters, sort, page, per_page, edit_id, tags.

POST /api/note#

Saves a plain text note. Reseek translates, tags, chunks, and embeds it through the same save pipeline used by the app.

Body: content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/link#

Saves a normal URL/link. Use the dedicated social and YouTube endpoints for richer extraction.

Body: url or content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/twitter#

Saves a Twitter/X status URL as Twitter content. Scraping and metadata extraction stay internal.

Body: url or content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/youtube#

Saves a YouTube URL as YouTube content. Transcript/metadata handling stays internal.

Body: url or content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/instagram#

Saves a public Instagram post or reel and extracts its caption, author, stats, and media.

Body: url or content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/facebook#

Saves a public Facebook page/group post or photo and extracts its content, author, stats, and media.

Body: url or content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/image#

Saves an image upload or image URL as image content. OCR stays internal.

Multipart body: file, or JSON body: url/content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/pdf#

Saves a PDF upload or PDF URL as PDF content. Text extraction stays internal.

Multipart body: file, or JSON body: url/content. Optional: description, metadata as JSON string, tags as JSON string array.

POST /api/video#

Saves a video upload or video URL as video content.

Multipart body: file, or JSON body: url/content. Optional: description, metadata as JSON string, tags as JSON string array.

GET /api/library/{id}#

Fetches one saved item by id.

Route param: id.

PATCH /api/library/{id}#

Changes one saved item. Only the fields you send change.

Write scope. Body: any of description, tags (array or JSON string array; an empty list removes all tags), content (notes only). Saved metadata and the original source stay as they are. If an edited note has Money entries, they are not changed; the note is flagged for review.

DELETE /api/library/{id}#

Permanently deletes one saved item and its uploaded file. This cannot be undone.

Write scope. Route param: id.

GET /api/tags#

Lists the tags used on your saved items.

Read scope. Query: optional q to match part of a tag name.

GET /api/stats#

How many items you have saved in total, today and this week, plus your most recent items.

Read scope. No body required.

Money#

GET /api/money/notes#

List saved notes needing money details.

Read scope. Use document_id to check one note, including completed extraction; page selects a result page.

POST /api/money/notes/resolve#

Resolve money details from a saved note.

Write scope. Use document_id, action and entry_index. Supply missing Money fields for record; ignore skips an entry; retry restarts a failed check; reviewed closes a completed review.

GET /api/money/owed#

Find money owed to you or by you, including partial repayments and settled entries.

Read scope. Query: query (person or description), direction (all/lent/borrowed), status (open/settled/voided/all), currency, page, limit. Totals cover all matches, separately by currency.

GET /api/money#

Search transactions with totals across all matches, grouped by currency.

Query: query (plain text, case-insensitive; quotes match a phrase), type, status, account_id, currency, payee, tag, date_from, date_to, min_amount, max_amount, sort, page, limit. Amount filtering/sorting requires currency.

GET /api/money/overview#

The Money page overview: income, spending, spending by tag, a trend and the previous period, combined into one display currency.

Read scope. Query: display_currency (required) plus the same filters as /api/money. Combined amounts are estimates at the returned rate date. When rates are unavailable, available is false and no combined total is returned.

GET /api/money/accounts#

List wallets, their recorded balances, defaults and supported currency precision.

No body required. Requires read permission.

POST /api/money/accounts#

Create an account or wallet.

Body: name, currency. Optional: opening_balance (decimal string), is_default.

PATCH /api/money/accounts/{accountId}#

Rename, archive/restore or choose a default account.

Body: name, is_default, archived. Currency and opening_balance cannot be changed.

POST /api/money/transactions#

Record expense, income, transfer, refund, lent, borrowed, received_back or paid_back. Both transfer legs are atomic.

Body: type, amount (decimal string), description, idempotency_key (or Idempotency-Key header). Supply account_id, or account_name + currency. Optional: occurred_on, payee, tags (array; an empty array means no tags), notes, document_id. Transfers: destination_account_id/name, destination_currency for new wallets, destination_amount for currency exchanges, fee. Refunds: optional refund_of_id when the original purchase is known; otherwise use a description. Lent/Borrowed: payee (person). Received back/Paid back: repayment_of_id (original loan UUID); positive amount in the original currency, up to the remaining amount.

GET /api/money/transactions/{transactionId}#

Get one transaction and its versioned history.

Route param: transactionId. Requires read permission.

PATCH /api/money/transactions/{transactionId}#

Correct a transaction and recalculate its balance effects.

Body: current version plus changed transaction fields. Reload on a version conflict.

DELETE /api/money/transactions/{transactionId}#

Void a transaction, reversing its balance effects while retaining history.

JSON body: current version. Void linked refunds before changing their original expense.

GET /api/money/quote#

Get a dated indicative exchange-rate quote. No money is moved.

Query: from, to (ISO currency codes), amount (decimal string). Returns estimated conversion and provider attribution; unavailable or stale rates return 422.

Reminders#

GET /api/reminders#

Lists reminders. Filter by status or search reminder content.

Query: optional status, query.

POST /api/reminders#

Creates a one-time or recurring email reminder.

Body: content, remind_at. Optional: timezone, frequency (once/minutely/hourly/daily/weekly/monthly/yearly), frequency_interval, weekday for weekly reminders, ends_at, document_id.

PATCH /api/reminders/{id}#

Edits, reschedules, pauses, or resumes a reminder.

Route param: id. Body: any reminder fields to change.

DELETE /api/reminders/{id}#

Cancels a reminder while preserving its delivery history.

Route param: id.

Quizzes#

GET /api/quizzes#

Lists grounded quizzes with each linked-source count and a small source preview.

Query: optional status and query. Search matches quiz titles and linked-source titles or summaries.

GET /api/quizzes/sources#

Lists saved items you can use as quiz sources, newest first.

Read scope. Query: optional query (matches titles and summaries), page. Returns 50 per page with has_more.

GET /api/quizzes/{id}/sources#

Lists every source currently linked to an owned quiz.

Route param: id.

POST /api/quizzes#

Creates a quiz grounded only in one or more saved items.

Body: title, document_ids (at least one; all selected items remain in one quiz), quiz_at. Optional: timezone, frequency (once/minutely/hourly/daily/weekly/monthly/yearly), frequency_interval, weekday for weekly quizzes, ends_at, question_mode, difficulty.

GET /api/quizzes/{id}/questions#

Lists delivered quiz questions and results newest first using cursor pagination.

Route param: id. Query: optional status (all/awaiting_answer/answered), per_page (1–100), cursor.

GET /api/quizzes/runs/{runId}#

Fetches one quiz question and its result. Correct answers remain hidden until answered.

Route param: runId.

PATCH /api/quizzes/{id}#

Edits, reschedules, pauses, resumes, or changes the source items of a quiz.

Route param: id. Body: any quiz fields to change. Use add_document_ids/remove_document_ids for incremental changes, or document_ids to replace the complete source set. At least one source must remain.

DELETE /api/quizzes/{id}#

Cancels a quiz while preserving its questions and results.

Route param: id.

POST /api/quizzes/runs/{runId}/answer#

Submits and grades one answer. Each quiz run accepts only one answer.

Route param: runId. Body: answer.